Document image skew detection based on the least distance fitting of the feature points
Jing Xie · Optical Technique · 2009
In process of document scanning input,document images inevitably introduce skew.While the algorithms for layout and character recognition are very sensitive to the page skew,so the skew detection and correction is an important step in the preprocessing for the document analysis.A new method of the document image skew detection is proposed based on the least distance fitting.First a series of image processing operations such as binarization,block partition,expansion and connected components labeling are made to extract the feature points.Then straight-line fitting of the feature points is done by the least distance method.Thus the skew angle of the document image is obtained.The experimental results show that the proposed method can detect the skew angle of the document image rapidly and accurately.